Skip to main content

Simulator

The simulator is a set of APIs in the SDK that can be used to test Wing applications without having to deploy the application to a cloud provider. The simulator works with any Wing application made with the SDK's cloud APIs.

Quick start guide​

Create a Wing simulator file​

To use the simulator, you will need to provide a Wing simulator file -- this is a file with a .wsim extension that is output when you compile your Wing program with the --target sim option. Check out the Wing Getting Started Guide for more information on how to write your first Wing program.

Using the simulator API (in Wing)​

🚧 Coming soon! 🚧

Using the simulator API (in TypeScript)​

Now, let's try starting the simulator and creating some resource clients to interact with the resources.

First, create an empty directory and add your .wsim file. Next, run npm install @winglang/sdk.

Let's create a file in the directory named main.ts:

import { testing } from '@winglang/sdk';

async function main() {
  const mySim = new testing.Simulator({ simfile: "hello.wsim" });
  await mySim.start();

  // (1)

  await mySim.stop();
}

void main();

Here, we create a new Simulator object, passing in the path to our simulator file, and then start the simulator. Inside (1) we can invoke methods on the simulator to get more information about the resources in our application. When we are done, we call sim.stop() to stop the simulator and clean up all resources (files, ports, etc.) that were created on your machine.

To print out a JSON tree view of the application, add the following line:

console.log(JSON.stringify(sim.tree, null, 2));

The tree contains the IDs of all the resources in the application, as well as details about dependencies between resources.

Now let's perform operations using a resource client. To obtain a resource's client, get the resource's path (from the JSON tree or elsewhere) and query the simulator with the getResource method. For example:

import { cloud } from '@winglang/sdk';

const fn = mySim.getResource("root/my_function") as cloud.IFunctionClient;
const response = await fn.invoke("hello!");
console.log(response);

Finally, when you want to understand how Wing resources are working, you may want to debug your application using traces and logs. Simulated resources automatically generate traces that you can inspect by calling listTraces() on your simulator instance.

console.log(JSON.stringify(mySim.listTraces(), null, 2));

Congratulations, you now know the ins and outs of using the Wing simulator! 🧑‍🎓

Check the API reference for more details about what methods are available on different resources and their inflight clients.